CAJ | 학술논문

实验室中的电子设备和化学试剂等对温度条件的要求较高，需要进行智能温度控制。传统的实验室温度控制方法采用 BP 神经网络控制方法，系统连接权值表现为一种静态属性相关权重，难以适应实验室温度自适应控制的需求。提出一种基于变结构 BP 神经网络自适应校正的实验室智能温度控制算法。进行实验室温度数据的挖掘和预处理，构建变结构 BP 神经网络自校正控制模型，采用自适应校正方法对温差进行调整，采用比例元进行温度过高情况下的微调，采用积分元进行温度过低下的微调，实现控制算法改进。基于 TMS320VC5509A DSP 芯片进行智能温控系统的核心电路设计。仿真结果表明，采用该系统能有效实现实验室温度智能控制，性能较好，可靠性高。
실험실중적전자설비화화학시제등대온도조건적요구교고，수요진행지능온도공제。전통적실험실온도공제방법채용 BP 신경망락공제방법，계통련접권치표현위일충정태속성상관권중，난이괄응실험실온도자괄응공제적수구。제출일충기우변결구 BP 신경망락자괄응교정적실험실지능온도공제산법。진행실험실온도수거적알굴화예처리，구건변결구 BP 신경망락자교정공제모형，채용자괄응교정방법대온차진행조정，채용비례원진행온도과고정황하적미조，채용적분원진행온도과저하적미조，실현공제산법개진。기우 TMS320VC5509A DSP 심편진행지능온공계통적핵심전로설계。방진결과표명，채용해계통능유효실현실험실온도지능공제，성능교호，가고성고。
The electronic equipment and chemical reagents in laboratory need intelligent temperature control due to their high requirements for temperature conditions. The traditional temperature control method with BP neural network control can not meet the requirements of laboratory temperature self?adaptive control since the system connection weight is a kind of relative weight of static attribute. An intelligent temperature control algorithm based on structure?transform BP neural network self?adap?tive correction is proposed for laboratories. On the basis of the data mining and preprocessing,the improvement of the control al?gorithm is realized by the methods that the BP neural network self?adaptive control model is constructed,the adaptive correction method is adopted to adjust the temperature difference,the proportion element is used to conduct fine adjustment of excess tem?perature,and the integral element is employed to execute fine adjustment of too low temperature. the core circuit of intelligent temperature control system based on DSP TMS320VC5509A chip was designed. The simulation results show that the system can effectively achieve the laboratory temperature intelligent control.